linux-kernel - Re: [LKP] [page cache] eb797a8ee0: vm-scalability.throughput -16.5% regression

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAHk-=wiXgigVGwumT0y_N64LjCiB5tR0snjCAm3u4YyhDB+dwQ@mail.gmail.com>
Date:   Wed, 27 Feb 2019 17:32:57 -0800
From:   Linus Torvalds <torvalds@...ux-foundation.org>
To:     "Huang, Ying" <ying.huang@...el.com>
Cc:     Waiman Long <longman@...hat.com>,
        Matthew Wilcox <willy@...radead.org>,
        "Chen, Rong A" <rong.a.chen@...el.com>, "lkp@...org" <lkp@...org>,
        LKML <linux-kernel@...r.kernel.org>,
        Andi Kleen <ak@...ux.intel.com>,
        Dave Hansen <dave.hansen@...el.com>,
        Tim C Chen <tim.c.chen@...el.com>
Subject: Re: [LKP] [page cache] eb797a8ee0: vm-scalability.throughput -16.5% regression

On Wed, Feb 27, 2019 at 5:19 PM Huang, Ying <ying.huang@...el.com> wrote:
>
> So I think in the heavily contended situation, we should put the fields
> accessed by rwsem holder in a different cache line of rwsem.  But in
> un-contended situation, we should put the fields accessed in rwsem
> holder and rwsem in the same cache line to reduce the cache footprint.
> The requirement of un-contended and heavily contended situation is
> contradicted.

Generally, we should strive to optimize for the uncontended state.

The performance profile of a contended state tends to be very
different, and the actual solution tends to be to try really hard to
just avoid contention to begin with.

I think we've gotten to the point where we have very few real loads
that show lock contention on a kernel level. And when people do find
loads that cause contention, we should try really hard to fix the
locking rather than try to then treat the symptom of contention.

So on the whole, aim to make the uncontended case go fast, at least to
a first approximation.

               Linus